From: Elijah Newren <newren@gmail.com>
To: Jeff King <peff@peff.net>
Cc: Michael J Gruber <git@grubix.eu>,
git@vger.kernel.org, Calvin Wan <calvinwan@google.com>
Subject: Re: [PATCH/RFD] fix connection via git protocol
Date: Mon, 17 Apr 2023 18:56:42 -0700 [thread overview]
Message-ID: <CABPp-BHW8Xi+=+GG4Z+-cf1a365eO6AzBifRjB+3600G5yyY9A@mail.gmail.com> (raw)
In-Reply-To: <20230417073827.GA39931@coredump.intra.peff.net>
On Mon, Apr 17, 2023 at 12:38 AM Jeff King <peff@peff.net> wrote:
>
> On Sat, Apr 15, 2023 at 10:47:35PM -0700, Elijah Newren wrote:
>
> > On Sat, Apr 15, 2023 at 4:06 AM Michael J Gruber <git@grubix.eu> wrote:
> > >
> > > 5579f44d2f ("treewide: remove unnecessary cache.h inclusion", 2023-04-11)
> > > broke connections via git protocol because it removed the inclusion of
> > > the default port macro. While some may consider this transport to be
> > > deprecated, it still serves some purpose.
> >
> > In particular the problem is that
> >
> > const char *port = STR(DEFAULT_GIT_PORT);
> >
> > translates now to
> >
> > const char *port = "DEFAULT_GIT_PORT";
> >
> > instead of
> >
> > const char *port = "9418";
> >
> > Since both compile and nothing in the testsuite tests this, I just
> > missed this problem when making the other changes.
>
> Your fix looks obviously correct, but the much more interesting thing to
> me is how surprising it is that neither the compiler nor tests caught
> it. The tests don't catch it because we never use the default port for
> our daemon tests, since we don't want two scripts running in parallel to
> conflict. And your example above shows what the compiler sees, but root
> issue is this funky string-ification macro:
>
> #define STR_(s) # s
> #define STR(s) STR_(s)
>
> The preprocessor doesn't know that we'll be confused if "s" isn't
> resolved, and by the time the compiler sees it, it's a string already.
>
> Obviously we could add a test that catches this at run-time, but we
> should be able to do better (catch it earlier, and with less code).
>
> My first thought was: why can't we just treat the port as an "int" in
> the first place? The answer is mostly that getaddrinfo() expects it as a
> string. It could even be a non-numeric service like "http" in theory
> (and looked up in /etc/services; Debian's even has "git" in it!), but
> our get_host_and_port() refuses to allow that. But even if we didn't
> want to ever support non-numeric service names, it makes the code more
> awkward (we have to format the port into an extra buffer).
>
> This would work:
>
> diff --git a/connect.c b/connect.c
> index fd3179e545..1eba71e34c 100644
> --- a/connect.c
> +++ b/connect.c
> @@ -753,7 +753,7 @@ static char *host_end(char **hoststart, int removebrackets)
> }
>
> #define STR_(s) # s
> -#define STR(s) STR_(s)
> +#define STR(s) (STR_(s) + BUILD_ASSERT_OR_ZERO(s))
>
> static void get_host_and_port(char **host, const char **port)
> {
>
> The error message is a bit verbose, but it starts with:
>
> connect.c: In function ‘git_tcp_connect_sock’:
> connect.c:801:32: error: ‘DEFAULT_GIT_PORT’ undeclared (first use in this function)
> 801 | const char *port = STR(DEFAULT_GIT_PORT);
> | ^~~~~~~~~~~~~~~~
>
> which seems OK in practice.
Seems pretty good to me.
> Another alternative is to just declare this STR() thing too clever, and
> put:
>
> #define DEFAULT_GIT_PORT_STR "9418"
>
> next to the int declaration. It's not like its going to change. But the
> BUILD_ASSERT doesn't seem too bad to me.
Yeah, I like the BUILD_ASSERT.
next prev parent reply other threads:[~2023-04-18 1:57 UTC|newest]
Thread overview: 101+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-01 15:10 [PATCH 00/24] Header cleanups (splitting up cache.h) Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 01/24] treewide: be explicit about dependence on trace.h & trace2.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 02/24] treewide: be explicit about dependence on advice.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 03/24] treewide: be explicit about dependence on convert.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 04/24] treewide: be explicit about dependence on pack-revindex.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 05/24] treewide: be explicit about dependence on oid-array.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 06/24] treewide: be explicit about dependence on mem-pool.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 07/24] treewide: remove unnecessary cache.h inclusion Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 08/24] object-name.h: move declarations for object-name.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 09/24] treewide: remove cache.h inclusion due to object-name.h changes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 10/24] git-zlib: move declarations for git-zlib functions from cache.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 11/24] treewide: remove cache.h inclusion due to git-zlib changes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 12/24] object-file.h: move declarations for object-file.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 13/24] treewide: remove cache.h inclusion due to object-file.h changes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 14/24] object.h: move an inline function and some defines from cache.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 15/24] editor: move editor-related functions and declarations into common file Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 16/24] treewide: remove cache.h inclusion due to editor.h changes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 17/24] pager.h: move declarations for pager.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 18/24] treewide: remove cache.h inclusion due to pager.h changes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 19/24] cache.h: remove unnecessary includes Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 20/24] strbuf: move forward declarations to beginning of file Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 21/24] treewide: remove double forward declaration of read_in_full Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 22/24] treewide: reduce includes of cache.h in other headers Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 23/24] chdir-notify, quote: replace cache.h include with path.h Elijah Newren via GitGitGadget
2023-04-01 15:10 ` [PATCH 24/24] mailmap, quote: move declarations of global vars to correct unit Elijah Newren via GitGitGadget
2023-04-03 16:23 ` [PATCH 00/24] Header cleanups (splitting up cache.h) Elijah Newren
2023-04-04 1:22 ` [PATCH v2 " Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 01/24] treewide: be explicit about dependence on trace.h & trace2.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 02/24] treewide: be explicit about dependence on advice.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 03/24] treewide: be explicit about dependence on convert.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 04/24] treewide: be explicit about dependence on pack-revindex.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 05/24] treewide: be explicit about dependence on oid-array.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 06/24] treewide: be explicit about dependence on mem-pool.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 07/24] treewide: remove unnecessary cache.h inclusion Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 08/24] object-name.h: move declarations for object-name.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 09/24] treewide: remove cache.h inclusion due to object-name.h changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 10/24] git-zlib: move declarations for git-zlib functions from cache.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 11/24] treewide: remove cache.h inclusion due to git-zlib changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 12/24] object-file.h: move declarations for object-file.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 13/24] treewide: remove cache.h inclusion due to object-file.h changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 14/24] object.h: move some inline functions and defines from cache.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 15/24] treewide: remove cache.h inclusion due to object.h changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 16/24] editor: move editor-related functions and declarations into common file Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 17/24] treewide: remove cache.h inclusion due to editor.h changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 18/24] pager.h: move declarations for pager.c functions from cache.h Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 19/24] treewide: remove cache.h inclusion due to pager.h changes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 20/24] cache.h: remove unnecessary includes Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 21/24] strbuf: move forward declarations to beginning of file Elijah Newren via GitGitGadget
2023-04-05 17:28 ` Calvin Wan
2023-04-07 6:07 ` Elijah Newren
2023-04-10 15:52 ` Calvin Wan
2023-04-10 16:47 ` Elijah Newren
2023-04-04 1:22 ` [PATCH v2 22/24] treewide: remove double forward declaration of read_in_full Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 23/24] treewide: reduce includes of cache.h in other headers Elijah Newren via GitGitGadget
2023-04-04 1:22 ` [PATCH v2 24/24] mailmap, quote: move declarations of global vars to correct unit Elijah Newren via GitGitGadget
2023-04-05 17:30 ` [PATCH v2 00/24] Header cleanups (splitting up cache.h) Calvin Wan
2023-04-07 7:08 ` Elijah Newren
2023-04-10 15:54 ` Calvin Wan
2023-04-11 3:00 ` [PATCH v3 00/23] " Elijah Newren via GitGitGadget
2023-04-11 3:00 ` [PATCH v3 01/23] treewide: be explicit about dependence on trace.h & trace2.h Elijah Newren via GitGitGadget
2023-04-11 3:00 ` [PATCH v3 02/23] treewide: be explicit about dependence on advice.h Elijah Newren via GitGitGadget
2023-04-11 3:00 ` [PATCH v3 03/23] treewide: be explicit about dependence on convert.h Elijah Newren via GitGitGadget
2023-04-11 3:00 ` [PATCH v3 04/23] treewide: be explicit about dependence on pack-revindex.h Elijah Newren via GitGitGadget
2023-04-11 3:00 ` [PATCH v3 05/23] treewide: be explicit about dependence on oid-array.h Elijah Newren via GitGitGadget
2023-04-11 7:41 ` [PATCH v3 06/23] treewide: be explicit about dependence on mem-pool.h Elijah Newren
2023-04-11 7:41 ` [PATCH v3 07/23] treewide: remove unnecessary cache.h inclusion Elijah Newren
2023-04-15 11:06 ` [PATCH/RFD] fix connection via git protocol Michael J Gruber
2023-04-16 5:47 ` Elijah Newren
2023-04-16 5:51 ` Elijah Newren
2023-04-16 11:21 ` Michael J Gruber
2023-04-17 7:38 ` Jeff King
2023-04-17 16:33 ` Junio C Hamano
2023-04-17 18:31 ` Junio C Hamano
2023-04-18 3:39 ` Jeff King
2023-04-18 20:47 ` Junio C Hamano
2023-04-18 1:56 ` Elijah Newren [this message]
2023-04-18 3:40 ` Jeff King
2023-04-17 16:29 ` Junio C Hamano
2023-04-18 1:55 ` Elijah Newren
2023-04-18 21:00 ` Junio C Hamano
2023-04-18 21:24 ` Eric Sunshine
2023-04-19 3:16 ` Elijah Newren
2023-04-24 20:57 ` Junio C Hamano
2023-04-19 3:15 ` Elijah Newren
2023-04-11 7:41 ` [PATCH v3 08/23] object-name.h: move declarations for object-name.c functions from cache.h Elijah Newren
2023-04-11 7:41 ` [PATCH v3 09/23] treewide: remove cache.h inclusion due to object-name.h changes Elijah Newren
2023-04-11 7:41 ` [PATCH v3 10/23] git-zlib: move declarations for git-zlib functions from cache.h Elijah Newren
2023-04-11 7:41 ` [PATCH v3 11/23] treewide: remove cache.h inclusion due to git-zlib changes Elijah Newren
2023-04-11 7:41 ` [PATCH v3 12/23] object-file.h: move declarations for object-file.c functions from cache.h Elijah Newren
2023-04-11 7:41 ` [PATCH v3 13/23] treewide: remove cache.h inclusion due to object-file.h changes Elijah Newren
2023-04-11 7:41 ` [PATCH v3 14/23] object.h: move some inline functions and defines from cache.h Elijah Newren
2023-04-11 7:41 ` [PATCH v3 15/23] treewide: remove cache.h inclusion due to object.h changes Elijah Newren
2023-04-11 7:41 ` [PATCH v3 16/23] editor: move editor-related functions and declarations into common file Elijah Newren
2023-04-11 7:41 ` [PATCH v3 17/23] treewide: remove cache.h inclusion due to editor.h changes Elijah Newren
2023-04-11 7:41 ` [PATCH v3 18/23] pager.h: move declarations for pager.c functions from cache.h Elijah Newren
2023-04-11 7:42 ` [PATCH v3 19/23] treewide: remove cache.h inclusion due to pager.h changes Elijah Newren
2023-04-11 7:42 ` [PATCH v3 20/23] cache.h: remove unnecessary includes Elijah Newren
2023-04-11 7:42 ` [PATCH v3 21/23] treewide: remove double forward declaration of read_in_full Elijah Newren
2023-04-11 7:42 ` [PATCH v3 22/23] treewide: reduce includes of cache.h in other headers Elijah Newren
2023-04-11 7:42 ` [PATCH v3 23/23] mailmap, quote: move declarations of global vars to correct unit Elijah Newren
2023-04-11 7:47 ` [PATCH v3 00/23] Header cleanups (splitting up cache.h) Elijah Newren
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CABPp-BHW8Xi+=+GG4Z+-cf1a365eO6AzBifRjB+3600G5yyY9A@mail.gmail.com' \
--to=newren@gmail.com \
--cc=calvinwan@google.com \
--cc=git@grubix.eu \
--cc=git@vger.kernel.org \
--cc=peff@peff.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).